Thirteen modules.
One data model.

114 verified capabilities behind these names — each checked against the product in continuous integration. Open any module for the full account.

Platform

• Design an approval workflow in a builder and attach it to any module
• Add your own fields without a developer
• License modules per organisation — buy one, add more later, with the product genuinely enforcing it
• Set permissions per role and override per user
• Run multiple facilities with their own work centres and transfers between them
• Migrate your data yourself: map columns, import, tie out the result, then go live
• Integrate over webhooks and a documented API rather than a bespoke connector project
• Enforce MFA and keep an audit log of who did what
• Run shop-floor workstations in a locked-down mode with their own session policy
• Export your data
• One source of truth that acts — an event in one module files dated, assigned work in another, automatically
• One tenant can never read or write another's data — isolation is enforced at the database on every write, not just in application code
• Every role sees exactly the modules it is granted and nothing more — verified role by role, with no privilege-escalation path
• AI is barred in code from the actions it must never take — it cannot post a journal entry, alter a Part 11 record, move a pipeline stage, or send an external message
• Multi-factor authentication is enforced on regulated workflows, backed by a complete, tamper-evident audit trail
• Every capability this site claims resolves to an inventory entry with evidence a reviewer can open, and the build fails when that evidence no longer exists
• Every structural check in the guard chain runs on every pull request — a guard wired into the chain but not into CI fails the build rather than reporting green from a job nobody invokes
• Known exceptions are held in a baseline whose count can only fall: a new one fails the build, and a fixed one keeps failing until it is removed from the baseline
• Forward demand is read through one path by every consumer of it — treasury coverage, margin ranking, the payment run, MRP and the constrained optimiser alike — so the org-grain predicate, the forward-source filter and the period window each exist exactly once
• Every read of the demand table must declare which grain it means, or the build fails — the check that stops a customer-level row and an org total being summed into one doubled figure
• Human-facing sequence numbers are minted in one place, and a new generator that counts rows instead of taking the maximum fails the build
• A second implementation of the forecast figure fails the build, against a baseline whose count can only fall